4. But there is the question of the linked light that must relate the visual organ to its object.
Now, firstly: since the intervening air is not necessary- unless in the purely accidental sense that air may be necessary to light- the light that acts as intermediate in vision will be unmodified: vision depends upon no modification whatever. This one intermediate, light, would seem to be necessary, but, unless light is corporeal, no intervening body is requisite: and we must remember that intervenient and borrowed light is essential not to seeing in general but to distant vision; the question whether light absolutely requires the presence of air we will discuss later. For the present one matter must occupy us:
If, in the act of vision, that linked light becomes ensouled, if the soul or mind permeates it and enters into union with it, as it does in its more inward acts such as understanding- which is what vision really is- then the intervening light is not a necessity: the process of seeing will be like that of touch; the visual faculty of the soul will perceive by the fact of having entered into the light; all that intervenes remains unaffected, serving simply as the field over which the vision ranges.
This brings up the question whether the sight is made active over its field by the sheer presence of a distance spread before it, or by the presence of a body of some kind within that distance.
If by the presence of such a body, then there will be vision though there be no intervenient; if the intervenient is the sole attractive agent, then we are forced to think of the visible object as being a Kind utterly without energy, performing no act. But so inactive a body cannot be: touch tells us that, for it does not merely announce that something is by and is touched: it is acted upon by the object so that it reports distinguishing qualities in it, qualities so effective that even at a distance touch itself would register them but for the accidental that it demands proximity.
We catch the heat of a fire just as soon as the intervening air does; no need to wait for it to be warmed: the denser body, in fact, takes in more warmth than the air has to give; in other words, the air transmits the heat but is not the source of our warmth.
When on the one side, that of the object, there is the power in any degree of an outgoing act, and on the other, that of the sight, the capability of being acted upon, surely the object needs no medium through which to be effective upon what it is fully equipped to affect: this would be needing not a help but a hindrance.
Or, again, consider the Dawn: there is no need that the light first flood the air and then come to us; the event is simultaneous to both: often, in fact, we see [in the distance] when the light is not as yet round our eyes at all but very far off, before, that is, the air has been acted upon: here we have vision without any modified intervenient, vision before the organ has received the light with which it is to be linked.
It is difficult to reconcile with this theory the fact of seeing stars or any fire by night.
If [as by the theory of an intervenient] the percipient mind or soul remains within itself and needs the light only as one might need a stick in the hand to touch something at a distance, then the perception will be a sort of tussle: the light must be conceived as something thrusting, something aimed at a mark, and similarly, the object, considered as an illuminated thing, must be conceived to be resistant; for this is the normal process in the case of contact by the agency of an intervenient.
Besides, even on this explanation, the mind must have previously been in contact with the object in the entire absence of intervenient; only if that has happened could contact through an intervenient bring knowledge, a knowledge by way of memory, and, even more emphatically, by way of reasoned comparison [ending in identification]: but this process of memory and comparison is excluded by the theory of first knowledge through the agency of a medium.
Finally, we may be told that the impinging light is modified by the thing to be seen and so becomes able to present something perceptible before the visual organ; but this simply brings us back to the theory of an intervenient changed midway by the object, an explanation whose difficulties we have already indicated.